the-high-council 0.7.7 → 0.7.9

package/.env.example

@@ -16,3 +16,8 @@ ZAI_API_KEY=
 # resumes with a higher ceiling. Unset means $7. Use --max-usd none to run
 # without a ceiling.
 MAX_USD_PER_RUN=7
+
+# Optional settings (README "Setup" lists them all): COUNCIL_WORKDIR and
+# COUNCIL_MAX_USD_LIMIT for the MCP server, COUNCIL_ALLOW_LOOPBACK_KEY_HOST=1
+# for a local proxy in front of a keyed provider, OLLAMA_API_KEY,
+# COUNCIL_PERSONAS_FILE, AUDIT_HMAC_KEY_FILE.

package/CONTRIBUTING.md

@@ -14,9 +14,9 @@ discussion first.
    in `node --test`, real calls are opt-in and cost real money.
 
 2. **A new chain config** (`chains/*.json`). A chain is data, not code - it names which seats run
-   which stages. `council doctor` validates any chain you add against every shipped-provider's key
-   presence and prices it worst-case with zero network calls; run that against your new file
-   before opening a PR. Look at `chains/plan-debate.json` or `chains/verify.json` as a starting
+   which stages. `council doctor --chain chains/<yours>.json` lints it (the same rules every run
+   applies, including the denied-model and `key-host` rules), and `council doctor` prices it
+   worst-case, both with zero network calls; run both against your new file before opening a PR. Look at `chains/plan-debate.json` or `chains/verify.json` as a starting
    shape depending on whether you want a debate-and-panel chain or a single-critic chain.
 
 3. **A new mock fixture** (`src/providers.js`'s `callMock`, exercised from `test/*.test.js`). If

package/PERSONAS.md

@@ -24,6 +24,10 @@ from the same public-domain legend and is not itself claimed as anyone's tradema
 here is a secondary association in a reader's mind, not a legal one - worth having on the record,
 not worth revisiting a decision already made.
 
+A `role.persona` key that is not in the set in use is sent as a bare name, with no voice
+directive. `plan-premium-7` seats two such keys, `hypatia` and `avicenna`, next to the five
+defaults: without a persona file of your own they reach the debate prompt as those names alone.
+
 **You can replace the whole set.** `loadPersonas()` in `src/personas.js` reads an operator-supplied
 JSON file (path via the `COUNCIL_PERSONAS_FILE` environment variable, or passed directly) and uses
 it in place of the default five entirely - nothing in this repo needs editing to do that. A

package/README.md

@@ -1,9 +1,16 @@
 # The High Council
 
-A planning harness that runs one idea past several AI models from different labs, makes them
-argue about software engineering, architecture, specific features, roadmaps and philosophy, and
-stops at a checkable result at an estimated price, with a variable number of debate rounds, model
-seats and labs (except xAI) and variable token limits per response and model.
+A planning harness that runs one idea past several AI models from different labs and makes
+them argue about software engineering, architecture, specific features, roadmaps and
+philosophy. It stops at a checkable result, at a price estimated before you spend anything.
+You choose the number of debate rounds, the seats and labs (any lab except xAI), and the token
+limit for each model's replies.
+
+We recommend our own chains: `cheap-7-v2`, `plan-premium-7` and `plan-highest-7` (the top
+models, GPT-6 Astra, Claude Opus 5.5, DeepSeek V4 Pro
+and GLM-5.3, write the alternative
+architectures and are the only seats that vote; cheaper seats write the proposals and the debate). You can also edit a chain to suit yourself, or run every seat on
+your own machine with `local-ollama`.
 
 See it work right now - no keys, no setup, no cost:
 
@@ -15,17 +22,48 @@ amended, a panel that splits in round 1 and signs off in round 2, and the handof
 model reply in it is written in advance; what is real is the code that runs the stages around
 them.
 
+**What it costs.** You pay the model providers directly, with your own API keys. The demo,
+`doctor`, `init` and the `mock` chains cost $0 and make no network call. `--dry-run` prints a
+run's worst-case price before anything is called, and every run stops before any stage that could
+take it past its spend cap ($7 unless you set `--max-usd`). Most runs cost less than the worst
+case: a panel that signs off early skips the remaining rounds.
+
+Three commands from nothing to a priced run of your own idea (no key needed until you drop
+`--dry-run`; [Your first ten minutes](#your-first-ten-minutes) walks through them):
+
+    npx the-high-council demo       # $0: every stage, scripted replies, no network
+    npx the-high-council init       # writes tasks/my-first-task.md for your idea, plus a $0 mock run
+    npx the-high-council --task tasks/my-first-task.md --chain cheap-7-v2 --dry-run
+
+**Which chain.** Worst-case prices as `council doctor` prints them, from the shipped price table
+(prices as of 2026-09-06) and the default task-size estimate; a long task costs more.
+
+| Chain | Who does what | Needs | Worst case |
+|---|---|---|---|
+| `cheap-7-v2` | Claude Sonnet 5 writes the plan; low-cost models from seven other labs review it, and all seven must sign off. Up to 7 rounds. | one OpenRouter key | $5.38 |
+| `plan-premium-7` | A Claude Code session writes the plan; a seven-lab panel of larger models (GPT-6 Astra, Claude Fable 5.1, Gemini 3.8 Flash, Muse Spark 1.2, Qwen 3.8 Max, DeepSeek V4 Pro, GLM-5.3) proposes, debates and reviews. | OpenRouter key + a Claude Code session | $22.30 |
+| `plan-highest-7` | Seven low-cost models propose and debate; four top models write whole architectures and are the only reviewers; one deep-dive seat checks the first draft against the task. **Untested with real models.** | OpenRouter key + a Claude Code session | $11.21 |
+| `local-ollama` | Every seat on your own machine, through Ollama. | Ollama and the models pulled | $0 |
+
+`plan-premium-7` and `plan-highest-7` cost more than the default cap, so give them a `--max-usd` at
+or above their dry-run price. Their writer seats are [external](#quick-start-mcp): the run pauses
+and a Claude Code session (or you) writes that stage, at no API cost. Nothing about any chain's
+output quality has been measured.
+
 Add it to Claude Code as an MCP server:
 
     claude mcp add council -- npx -y the-high-council council --mcp
 
+It is also listed in the official MCP Registry as `io.github.muad-yasin/the-high-council`, and
+there is a one-file Claude Desktop bundle ([Quick start (MCP)](#quick-start-mcp)).
+
 Runs as an **MCP server** (so an agent like Claude Code can drive it) or as a **CLI**. Bring your
 own API keys. Nothing is resold, nothing is hosted for you, and your keys go only to the
 providers you choose, never to us.
 
 Run `npx the-high-council council doctor` to see which of your own keys are
 set, which shipped chains you can already run with them, and what each would cost - before
-spending anything.
+spending anything. Where it falls short: [Known limits](#known-limits-stated-plainly).
 
 > **On what this does and doesn't claim.** This repo publishes the mechanism: the chains, the seat
 > rosters, the stage order. It does not claim to produce better plans than a single good model
@@ -112,6 +150,10 @@ A run moves through fixed stages. Which stages fire depends on the chain you pic
    again, up to the chain's round cap.
 8. **Handoff** - a `HANDOFF.md` written for whoever executes the result.
 
+Some chains add optional stages to this: whole alternative architectures written blind before
+the skeleton, a deep-dive check of the first draft, a dispute stage that records unresolved
+objections at the top of the plan, and a final security review.
+
 Every run writes a folder: the deliverable, `BOARD.md` (the full debate - every post, every
 withdrawal), `HANDOFF.md`, a per-lab scoreboard, and the real token/cost accounting.
 
@@ -124,7 +166,7 @@ which are experimental, and how versions change: [docs/report-format.md](docs/re
 
 ```jsonc
 {
-  "schemaVersion": 1, "runId": "...", "chain": "...", "passed": false, "maxUsd": 5,
+  "schemaVersion": 1, "runId": "...", "chain": "...", "passed": false, "maxUsd": 7,
   "criteria":  [ "each acceptance criterion, as written before the debate" ],
   "questions": [ { "question": "...", "why": "...", "default": "..." } ],  // null if the chain skipped them
   "proposals": [ { "id": "DEEPSEEK-1", "lab": "deepseek", "model": "...",
@@ -136,7 +178,11 @@ which are experimental, and how versions change: [docs/report-format.md](docs/re
     "posts":   [ { "by": "deepseek", "on": "QWEN-1", "stance": "object|support|merge",
                    "text": "...", "merge_with": "GLM-2" } ],
     "replies": [ { "id": "DEEPSEEK-1", "action": "keep|amend|withdraw",
-                   "text": "...", "replaced_by": "GLM-1" } ]
+                   "text": "...", "replaced_by": "GLM-1" } ],
+    // Since 0.7.8 (experimental): posts and replies the harness rejected, counted per lab and
+    // reason (unknown_target, own_proposal, bad_stance, not_own_proposal, not_posted_on,
+    // bad_action, unreadable). [] when nothing was dropped.
+    "dropped": [ { "stage": "debate|replies", "by": "qwen", "reason": "unknown_target", "count": 1 } ]
   },
   // Only on chains with `alternatives: { enabled: true }`: one whole architecture per lab, written
   // blind before the skeleton, and the same post/reply board the proposals get.
@@ -189,6 +235,11 @@ those plus `text` and may carry more (an `amend` often brings a `how`). `replace
 when the action is `withdraw`. `proposals[]` likewise carries `withdrawn` / `amended` /
 `replaced_by` only when they apply. Read defensively: check for a field, do not assume it.
 
+Chains with newer, opt-in stages add their own fields (`deep_dive`, `notQuorate`, `dispute`,
+`security_review`, ...); [docs/report-format.md](docs/report-format.md) lists which are stable and
+which experimental. With the majority guard on, a withdrawal that does not quote the argument it
+concedes to carries `unargued: true` in `debate.replies[]` and its proposal stays for the builder.
+
 `signoff[].lab` is the seat's lab; `signoff[].provider` holds the same value under an older,
 misleading name and is deprecated (kept until the report format's version 2).
 `signoff[].objections` is why a seat declined, on the seat's own record. `signedOff: null` means
@@ -225,13 +276,30 @@ ones. Put only the keys your chosen chain needs in that `.env`, one `NAME=value`
 `DEEPSEEK_API_KEY` · `GROQ_API_KEY` · `TOGETHER_API_KEY` · `COHERE_API_KEY` ·
 `OPENROUTER_API_KEY` · `ZAI_API_KEY`
 
-xAI/Grok and Kimi/Moonshot are never seated, and neither is a router id (`openrouter/auto`, the
-Pareto Code router) that could route to them: chain-lint refuses such a chain and the run itself
-refuses such a seat, with no override. There is no `xai` provider.
+xAI/Grok is never seated, and neither is a router id (`openrouter/auto`, the Pareto Code router)
+that could route to it: chain-lint refuses such a chain and the run itself refuses such a seat,
+with no override. There is no `xai` provider. Kimi/Moonshot models are not in any chain this
+package ships, but you may seat one in a chain of your own.
 
 The CLI refuses to start if any seat in the chosen chain is missing its key, rather than failing
 halfway through a paid run. `ollama` (below) is the one exception - it needs no key at all.
 
+A keyed seat sends its key only to its own provider's https API. A chain that points a keyed
+seat's `baseUrl` anywhere else is refused before any call (see
+[Local models](#local-models-ollama-lm-studio-)).
+
+Other settings, all optional, read from the environment or `.env`:
+
+| Variable | What it does |
+|---|---|
+| `MAX_USD_PER_RUN` | The default per-run spend cap (`7` when unset). `--max-usd` overrides it for one run. |
+| `COUNCIL_WORKDIR` | MCP server only: the folder for `tasks/` and `runs/`, when the client starts the server somewhere else. |
+| `COUNCIL_MAX_USD_LIMIT` | MCP server only: a cap the MCP client cannot raise or remove through `max_usd`. |
+| `COUNCIL_ALLOW_LOOPBACK_KEY_HOST` | `1` lets a keyed seat use a loopback `baseUrl` (a local proxy in front of the provider). Nothing else. |
+| `OLLAMA_API_KEY` | Sent to an `ollama` seat's server only if you set it. |
+| `COUNCIL_PERSONAS_FILE` | Your own persona set for debate seats ([PERSONAS.md](PERSONAS.md)). |
+| `AUDIT_HMAC_KEY` / `AUDIT_HMAC_KEY_FILE` | Signs the opt-in audit log (see [Privacy](#privacy)). |
+
 ### From source
 
 ```bash
@@ -279,6 +347,12 @@ If your local runner listens somewhere else (LM Studio, a different port, a remo
 { "provider": "ollama", "model": "mistral", "baseUrl": "http://localhost:1234/v1" }
 ```
 
+Put a local or self-hosted server on an `ollama` seat, as above. A seat with a keyed provider
+(`openrouter`, `openai`, ...) sends that provider's key to its `baseUrl`, so chain lint refuses
+one whose `baseUrl` is anything but that provider's own https API address (`key-host`), and the
+CLI lints before every run and every resume. For a local proxy in front of the provider, set
+`COUNCIL_ALLOW_LOOPBACK_KEY_HOST=1` in your environment: it allows a loopback address, nothing else.
+
 A local seat prices at $0 in every estimate and run report - `dry_run` shows it as priced, not
 "unpriced," and it can never trip the spend cap. See `chains/local-ollama.json` for a full
 example chain with every seat local. As with every other seat in this project: BYOK is really
@@ -318,6 +392,16 @@ will not accidentally publish them.
 `verify` is the default when you name no chain: a panel of two labs, a hard two-round cap. It
 needs three keys (Anthropic, OpenAI and Google, one account each). With one OpenRouter key, start
 with `cheap-7-v2` instead; `council doctor` lists every chain that runs on the keys you have.
+`--chain` takes a chain's name (`cheap-7-v2`), never a file path: a `chains/<name>.json` of your
+own in the working directory is found by its name.
+
+**If your task names files.** A task that names a file (`src/app.js`) without including its
+content is stopped before any call (exit 9, `BLOCKED-ARTIFACTS.md`), because the seats would
+otherwise plan against code they never saw. Put the real content in with
+`council fence --task tasks/x.md --repo <path to your repo> src/app.js` (it appends each file,
+fenced and labelled, and refuses to write if it finds a key), read the task file, then run.
+If a name is only a location, pass `--allow-unfenced` for the whole task or
+`--allow-unfenced app.js` for named files. Everything in the task file goes to every seat.
 
 ### Exit codes
 
@@ -327,7 +411,7 @@ Each outcome has its own code, so a script or CI job can branch on it:
 |---|---|
 | 0 | Finished: a deliverable and `report.json` were written. |
 | 1 | The chain failed lint, or an input is missing (no task file, a resume from the wrong directory). |
-| 2 | Usage error: a flag is missing its value or has an invalid one. |
+| 2 | Usage error: a flag is missing its value or has an invalid one, `--chain` is not a chain name, a `--context` file is not a regular file or is over 2 MB, or `--rematch`/`--replay` of a run given `--criteria`. |
 | 3 | Paused at an external seat: answer `NEEDS-<stage>.md`, then `--resume`. |
 | 4 | Stopped at the per-run spend cap before the next stage was paid for; resume with a higher `--max-usd`. |
 | 5 | A provider key the chain needs is missing (degradable: set it and run again). |
@@ -336,7 +420,7 @@ Each outcome has its own code, so a script or CI job can branch on it:
 | 8 | Finished, but the final security review could not judge it (never a pass). |
 | 9 | Artifact gate: the task names files whose content it never includes (`BLOCKED-ARTIFACTS.md`). |
 | 10 | Preflight: a seat objected to the task description itself (`STOPPED-preflight.md`). |
-| 11 | PII gate (`--pii-gate hard-stop`): the input matches a PII shape or one of the key formats listed in `src/secret-patterns.js`. |
+| 11 | Something in the input must not be sent: a prompt matches one of the key formats listed in `src/secret-patterns.js` (checked by default; `--allow-secret-shaped` overrides), or, under `--pii-gate hard-stop`, a PII shape. |
 | 12 | `policy.json` refused the chain. |
 | 13 | Another process is already running this run folder. |
 | 14 | The task file changed since the run started and `AMENDMENTS.md` does not cover it. |
@@ -377,14 +461,16 @@ than reporting the overspend afterwards.
 
 ```bash
 council --task tasks/x.md --max-usd 2     # this run stops at $2
-council --task tasks/x.md --max-usd none  # no ceiling
+council --task tasks/x.md --max-usd none  # no ceiling (0 is refused: it used to mean this)
 export MAX_USD_PER_RUN=20                         # change the default
 ```
 
 A run that hits the ceiling stops cleanly and writes `STOPPED-budget.md` saying what it spent, which
 stage it stopped at, and what that stage would have cost. Nothing is half-written: it produces no
-`deliverable.md` and no `report.json`, so a stopped run never reads as a finished one. Continue it
-with a higher ceiling - completed stages replay from disk and cost nothing the second time:
+`deliverable.md` and no `report.json`, so a stopped run never reads as a finished one. What it did
+get through is in `report-partial.json` and `BOARD-partial.md`
+([docs/report-format.md](docs/report-format.md)). Continue it with a higher ceiling - completed
+stages replay from disk and cost nothing the second time:
 
 ```bash
 council --resume runs/<id> --max-usd 10
@@ -463,6 +549,17 @@ up in `claude mcp list` as `plugin:the-high-council:high-council`. Remove it wit
 local clone instead (`claude plugin marketplace add /path/to/clone`), the server runs from the
 clone itself, so run `npm install` in it first.
 
+**Claude Desktop:** download `the-high-council-<version>.mcpb` from the
+[GitHub Releases](https://github.com/muad-yasin/the-high-council-mcp/releases) page (its SHA-256
+is next to it) and open it. Claude Desktop shows a settings form: a council folder for tasks and
+runs, a per-run spend ceiling that Claude cannot raise or remove from a chat, and one optional
+field per provider key (marked sensitive, so the app keeps them in the OS keychain). The bundle is unsigned. How it is built:
+[mcpb/README.md](mcpb/README.md).
+
+**MCP Registry:** the server is listed in the official registry as
+`io.github.muad-yasin/the-high-council` (npm package, stdio). A client that installs from the
+registry runs the same `npx` command as above.
+
 Registering with a client other than Claude Code (DeepSeek Harness, OpenHands, Cline, goose,
 Continue): [docs/mcp-clients.md](docs/mcp-clients.md) has the real config for each.
 
@@ -472,6 +569,13 @@ own takes precedence over the bundled ones. In a git project, add `.env` (and `r
 ideas are private) to `.gitignore` before an agent commits everything; `council doctor` warns
 when `.env` is not ignored.
 
+The server only works inside that directory. `dry_run` and `start_run` take a chain by name.
+Every path `start_run` is given (`task`, `draft`, `context`, `from_run`, and each file inside a
+`context` folder or `from_run` run folder) must resolve inside the working directory, after
+symlinks, and must not be on the secret/credential denylist; outside `tasks/`, `runs/` and
+`context/` a gitignored file is refused too. Set `COUNCIL_MAX_USD_LIMIT` in the server's
+environment to cap what the client's `max_usd` can ask for.
+
 To run it by hand:
 
 ```bash
@@ -485,18 +589,18 @@ Then drive it with these tools:
 | `list_chains` | Available chains with descriptions and worst-case price |
 | `write_task` | Write the request the harness plans against |
 | `dry_run` | Price a run before spending anything |
-| `start_run` | Start a run in the background (`max_usd` sets its ceiling) |
+| `start_run` | Start a run in the background (`max_usd` sets its ceiling). `chain` is a chain name; `task`, `draft`, `context` and `from_run` must be inside the working directory |
 | `run_status` | Stage reached, panel verdicts, scoreboard, cost, budget remaining |
 | `spend_report` | What every run has cost across a window of days, plus `session_cost_today` (calendar-day, per-model breakdown), derived from disk |
 | `verdict_stats` | How the debate mechanism itself is doing, per chain and per lab: sign-off rate, rounds, objections, dropouts, cost/wall time, largest prompt files - derived from disk |
 | `metrics_report` | Descriptive telemetry only (never an evaluation or baseline): amendment rate, withdrawal rate, objection-follow-through rate, tool-call usage, consensus-induced-regression count - derived from disk. See [`docs/paid-comparison-run.md`](docs/paid-comparison-run.md) for what a *real* comparison against a single model would need - not built, not this metric |
-| `external_prompt` | The prompt a paused external stage is waiting on - **read it with this, not by opening the file**; the prompts are routinely tens of thousands of tokens and most clients silently truncate a file read |
+| `external_prompt` | The prompt a paused external stage is waiting on - **read it with this, not by opening the file**; the prompts are routinely tens of thousands of tokens and most clients silently truncate a file read. `waiting` lists every stage the run waits on (a panel of external critics pauses together); pass `stage` for another one's prompt |
 | `prepare_stage_prompt` | Write a self-contained bundle for a paused stage, so a driving session doing other work at the same time can dispatch it to a fresh subagent instead of authoring it inline - see [`docs/dispatch-pattern.md`](docs/dispatch-pattern.md) |
-| `submit_stage` | Answer a paused stage and resume. Optional `claimed_by` records who's answering; a second submission for an already-answered stage no longer errors - it's kept as `<stage>.late.md` with a warning instead of being rejected |
+| `submit_stage` | Answer a paused stage; the run resumes once every stage it waits on has an answer. Optional `claimed_by` records who's answering; a second submission for an already-answered stage no longer errors - it's kept as `<stage>.late.md` with a warning instead of being rejected |
 | `resume_run` | Resume a paused run, or raise `max_usd` on one the cap stopped |
-| `read_run_file` | Read any file from a run folder |
+| `read_run_file` | Read any file from a run folder (`report-partial.json` for a run the cap stopped) |
 | `list_runs` | Past runs |
-| `plan_outline` | Outline pass |
+| `plan_outline` | Section tree of a run's deliverable (or a `.md` file in the working directory) with word counts |
 
 **External vs. API-backed seats.** `{ "provider": "external" }` pauses a run at that seat so a
 Claude Code session on a flat-rate subscription plays it - free at the point of use, since the
@@ -509,13 +613,23 @@ for anyone with a subscription, since the project's cost story depends on it.
 
 ## Chains
 
-49 chain configs live in `chains/`. Each one is plain JSON - the seat roster, which models fill
+52 chain configs live in `chains/`. Each one is plain JSON - the seat roster, which models fill
 which seat, the round cap, and whether proposals/debate/handoff stages run. They are meant to be
 copied and edited.
 
-A few worth knowing:
-
-- **`verify`** - two labs, two rounds. The recommended starting point.
+The four recommended ones (`cheap-7-v2`, `plan-premium-7`, `plan-highest-7`, `local-ollama`) are
+in the table at the top. A few more worth knowing:
+
+- **`verify`** - two labs, two rounds. The default when you name no chain; needs Anthropic, OpenAI
+  and Google keys.
+- **`plan-open-7`** - the `cheap-7-v2` panel, with a Claude Code session writing the plan instead
+  of a paid seat.
+- **`plan-daily-7`** - `plan-highest-7`'s stages at a lower price, for planning one feature:
+  Claude Opus 5.5 on your Claude subscription (an external seat), GPT-6 Luna and DeepSeek V4 Pro
+  write the architectures and alone vote; seven low-cost labs propose and debate. Two Claude Code
+  sessions answer its external seats: Opus 5.5 every stage whose label contains `opus5.5-sub`
+  (including `-retry` and `-reask1`/`-reask2` forms), Sonnet 5 the writer stages. $1.90 worst case.
+  **Run once with real models (2026-09-28); nothing about it has been measured.**
 - **`cheap`** - small models throughout. For testing the harness itself, not for real work.
 - **`plan-debate`** - five labs propose blind, debate each other's proposals anonymised, then a
   blind panel grades the integrated draft.
@@ -544,6 +658,17 @@ Read a chain's `description` field before running it; they say what they cost yo
   source-level guard, not just an intention. A chain with no `role` set is byte-identical to
   before. **No efficacy claim**: this ships the mechanism, not evidence that it helps - see
   CHANGELOG.md's 0.6.0 entry.
+- **`quorum: { "minHeard": N }`** - opt-in, unanimous chains only, set in no shipped chain. A
+  round's sign-off counts only when at least N reviewers gave a verdict (signed off or objected);
+  a round short of it is recorded as `notQuorate`, the run does not pass, and the loop stops.
+- **Tiered councils (experimental, used by `plan-highest-7`, `plan-daily-7` and `mock-tiered`).**
+  `seats.alternatives` names who writes the whole alternative architectures, separately from the
+  proposers. `deep_dive` + `seats.deep_dive` add one non-voting seat with its own dollar cap
+  (`usd`, required) inside the run's cap. `majority_guard.enabled` shows each author every
+  argument once, with no lab name or count, and keeps a proposal whose withdrawal does not quote
+  the argument it gives in to. chain-lint refuses a tiered chain where a lab or model sits
+  both among the proposers and among the reviewers or architecture authors, or where the deep-dive seat shares a lab or model with a reviewer. See
+  CHANGELOG.md's 0.7.8 entry.
 
 ## How this was built
 
@@ -618,8 +743,16 @@ what each rule guards against and why. Copy the folders you want into your proje
   such a reply is recorded as `REASONING_EXHAUSTED` and is not retried: the seat abstains, which
   blocks unanimity and never counts as consent. Whether to change a chain's reasoning settings for it
   is a roster decision, not something the harness does for you.
-- Prices in `src/pricing.json` are hand-maintained list prices, last verified 2026-09-06. They are
-  estimates, not invoices. Your provider's bill is the real number.
+- Prices in `src/pricing.json` are hand-maintained list prices. Its `asOf` date (2026-09-06: the
+  oldest date every entry was last checked) is printed by `council doctor` and `--dry-run`, with a
+  warning once it is more than 60 days old. The spend cap and every estimate use these static
+  prices. They are estimates, not invoices. Your provider's bill is the real number.
+- `plan-highest-7` has not been run with real models, and `plan-daily-7` has run once
+  (2026-09-28). None of the tiered councils' mechanisms (tiers, the deep-dive seat, the majority
+  guard) has been measured. Their dry-run prices are projections.
+- Seats marked `external` (the writer seats in `plan-premium-7`, `plan-highest-7`,
+  `plan-daily-7` and `plan-open-7`, and the Opus 5.5 anchor seat in `plan-daily-7`) pause the run until someone answers them, and are $0 only because a Claude Code
+  session on a subscription writes them. Nothing checks who wrote the answer.
 - Chains with many labs and high round caps get expensive quickly. `plan-unanimous` at three rounds
   is fifteen critic calls; `plan-auto` runs several frontier models over multiple rounds. Price
   before you run.
@@ -632,6 +765,10 @@ what each rule guards against and why. Copy the folders you want into your proje
   which is the default, it does not detect a full rewrite**: anyone who can edit the file can
   recompute every hash. Set `AUDIT_HMAC_KEY_FILE` to a key kept outside the run folder if the log
   has to stand up to that.
+- The outbound key scan matches **key shapes only**, the formats in `src/secret-patterns.js`; a
+  password written in prose is not detectable. Generic shapes (a password inside a URL or an
+  assignment) are checked in your inputs before the run; text the models write is checked only for
+  distinctive key formats.
 
 ## Privacy
 
@@ -669,8 +806,9 @@ hand, 2026-09-13, on the live site:
 - No custom widget on any page (`index.html`, `board.html`, `demo.html`) uses a `<div>` or
   `<span>` with a click handler in place of a real interactive element - checked directly in the
   page source, not inferred.
-- No image ships anywhere in `docs/` (checked directly - zero `<img>` tags across all three
-  pages), so there is no informative image that could be missing alt text.
+- The only image in `docs/` is the local logo (`logo-diamond.png`): the header copy carries alt
+  text, the faint background copy on `index.html` has an empty `alt=""` as decoration.
+  `test/landing-page.test.js` checks both on every page.
 - Primary body text (`#8d8a84`/`#b9b5ad`) against the page background (`#0a0a0b`) measures
   roughly 5.7:1 contrast by the WCAG relative-luminance formula, above the 4.5:1 AA threshold
   for normal text.

package/TROUBLESHOOTING.md

@@ -96,8 +96,28 @@ local chains as runnable because they need no key, not because the server is up.
 ### A run stopped with `STOPPED: per-run spend cap reached` (exit 4)
 
 The next stage's worst case would have crossed the cap. Nothing past the cap was spent, and no
-plan was written yet. `council --resume runs/<id> --max-usd <higher>` continues from where it
-stopped. The run tells you at the start when a chain's worst case is above your cap.
+plan was written yet; what the run got through is in `report-partial.json` and `BOARD-partial.md`.
+`council --resume runs/<id> --max-usd <higher>` continues from where it stopped. The run tells you
+at the start when a chain's worst case is above your cap. `plan-premium-7` and `plan-highest-7`
+are above the default $7 cap before they start: price them with `--dry-run` and pass a
+`--max-usd` at or above that.
+
+### `BLOCKED: ... file(s) named but never fenced` (exit 9)
+
+The task names a file (`src/app.js`, `package.json`) but does not include its content, so the
+seats would plan against code they never saw. Nothing was called. Either append the real files
+with `council fence --task tasks/x.md --repo <path> <file> ...` and read the task file before
+running, or, if the names are only locations, run with `--allow-unfenced` (whole task) or
+`--allow-unfenced a.js,b.ts` (those files). `BLOCKED-ARTIFACTS.md` in the run folder lists the
+names it found.
+
+### `--chain` refused, or a chain lint error naming `key-host`
+
+`--chain` takes a chain's name (`cheap-7-v2`, or `my-chain` for your own `chains/my-chain.json`),
+never a path; anything else is a usage error (exit 2). `key-host` means a seat with a keyed
+provider has a `baseUrl` that is not that provider's own https API, and its key would go there.
+Move a local or self-hosted server to an `ollama` seat, or, for a local proxy in front of the
+provider, set `COUNCIL_ALLOW_LOOPBACK_KEY_HOST=1` (loopback addresses only).
 
 ### `.env holds your API keys and git would commit it`
 

package/chains/mock-tiered.json

@@ -0,0 +1,43 @@
+{
+  "name": "mock-tiered",
+  "description": "Offline. A tiered council at $0 (tiered councils, 2026-09-26): mass seats (seats.proposers) propose parts, debate them and post on the architectures; anchor seats (seats.alternatives) write the whole architectures and reply; the anchor panel (seats.critics) alone votes. The majority guard is on: authors see arguments without lab letters or counts, and a withdrawal that quotes no argument (mock-yield-unargued, mock-alt-withdraw) stays for the builder while one that does (mock-yield-argued) is honoured. A deep-dive seat reads the task in 1000-character chunks, once per focus entry, under its own cap and maxCalls, and the reviser answers its findings before the panel's first review.",
+  "maxRounds": 2,
+  "signoff": "unanimous",
+  "dispute": { "enabled": true, "stall_rounds": 2 },
+  "proposals": { "parts": 2, "maxTokens": 400 },
+  "debate": true,
+  "decisions": { "enabled": true },
+  "alternatives": { "enabled": true, "debaters": "all" },
+  "majority_guard": { "enabled": true },
+  "canary": { "enabled": true, "sampleRate": 1 },
+  "deep_dive": {
+    "enabled": true,
+    "job": "sources",
+    "focus": ["Every requirement in the excerpt.", "Every number in the excerpt."],
+    "usd": 1,
+    "maxCalls": 5,
+    "chunkChars": 1000
+  },
+  "handoff": true,
+  "estimate": { "promptTokens": 1000, "draftTokens": 1000, "critiqueTokens": 300 },
+  "seats": {
+    "criteria": { "provider": "mock", "model": "mock-criteria" },
+    "skeleton": { "provider": "mock", "model": "mock-skeleton" },
+    "builder": { "provider": "mock", "model": "mock-builder" },
+    "reviser": { "provider": "mock", "model": "mock-builder" },
+    "proposers": [
+      { "provider": "mock", "model": "mock-yield-argued", "lab": "mass-a" },
+      { "provider": "mock", "model": "mock-yield-unargued", "lab": "mass-b" },
+      { "provider": "mock", "model": "mock-proposer-a", "lab": "mass-c" }
+    ],
+    "alternatives": [
+      { "provider": "mock", "model": "mock-alt-a", "lab": "anchor-a" },
+      { "provider": "mock", "model": "mock-alt-withdraw", "lab": "anchor-b" }
+    ],
+    "critics": [
+      { "provider": "mock", "model": "mock-critic-a", "lab": "anchor-a" },
+      { "provider": "mock", "model": "mock-critic-b", "lab": "anchor-b" }
+    ],
+    "deep_dive": { "provider": "mock", "model": "mock-deep", "lab": "deep-a", "maxTokens": 128000 }
+  }
+}

package/chains/plan-daily-7.json

@@ -0,0 +1,43 @@
+{
+  "name": "plan-daily-7",
+  "description": "daily7 (the owner's name, 2026-09-27: cheap API seats, and the expensive Opus 5.5 seat runs on a local Claude Code session). Tiered council for planning ONE feature cheaply (Muad, 2026-09-27: \"Opus 5.5 on a local sub that does the important tasks from the highest7 setup ... all together ~1$ per council run\"). Same stages and flags as plan-highest-7 (alternatives, debate, decision records, deep dive, majority guard ON, canary, dispute, handoff); only the seats and the deep-dive cap differ. ANCHOR SEATS write the whole alternative architectures (seats.alternatives), answer the posts on them, and are the only panel that votes, unanimously (seats.critics): Claude Opus 5.5 on the owner's Claude subscription (provider external, model subscription:opus-5.5, lab opus5.5-sub: $0 API, a separate Opus 5.5 Claude Code session answers its paused stages), GPT-6 Luna and DeepSeek V4 Pro through OpenRouter. MASS SEATS (seats.proposers, seven labs from the strong and low-cost part of arena.ai's agent leaderboard, fetched 2026-09-27): Tencent Hy4 preview, Gemini 3.8 Flash, DeepSeek V4.1 Flash, GLM 5.2, Muse Spark 1.3, Qwen 3.8 Max, Mistral Medium 3.5 (kept on the owner's word) write the blind proposals and every debate post and reply, and post on the anchors' architectures (alternatives.debaters \"all\"); they argue, they never vote. DEEP-DIVE SEAT: DeepSeek V4.1 Flash, a mass-tier model (a different model from the voting V4 Pro, so chain-lint deep-dive-votes passes), under its own $0.50 cap: a feature-sized task is one 40,000-character call, and a $4 sub-cap would be larger than this chain's whole worst case. Criteria: Llama 4 Maverick, which holds no other seat. WRITER SEATS (skeleton, build, deep-dive-revise, revise-N, dispute, handoff) are a Claude Code session running Sonnet 5, which builds and holds no panel seat. The Opus 5.5 panel seat is a DIFFERENT Anthropic model in a separate session: the owner's plan-highest-7 exception (an Anthropic model reviews a draft another Claude writes), not selfReview - no model writes and votes. ROUTING, because a NEEDS file names only its label: every label containing opus5.5-sub goes to the Opus 5.5 session (alternative-, alt-debate-, alt-reply-, panel-N-, including the suffixed forms panel-N-opus5.5-sub-retry, -reask1, -reask2 and alternative-opus5.5-sub-retry); skeleton, build, deep-dive-revise, revise-N, dispute and handoff go to the Sonnet 5 session. Nothing checks who wrote an answer. Dry run on a one-page feature task (2026-09-27, GPT-6 Luna in place of GPT-6 Sol on the owner's word: \"GPT 6 Sol is too expensive, take GPT 6 luna\"): $1.90 worst case at 7 rounds; under the default $7 cap. No Kimi, no xAI. Run once with real models (2026-09-28); none of its mechanisms has been measured. Price your task with `npm run dry` and get an explicit go before any paid run.",
+  "maxRounds": 7,
+  "signoff": "unanimous",
+  "dispute": { "enabled": true, "stall_rounds": 2 },
+  "proposals": { "parts": 3, "maxTokens": 4000 },
+  "debate": true,
+  "decisions": { "enabled": true },
+  "alternatives": { "enabled": true, "debaters": "all" },
+  "deep_dive": { "enabled": true, "job": "sources", "usd": 0.5, "maxCalls": 24, "chunkChars": 40000 },
+  "majority_guard": { "enabled": true },
+  "canary": { "enabled": true, "sampleRate": 0.1 },
+  "handoff": true,
+  "estimate": { "promptTokens": 10000, "draftTokens": 10000, "critiqueTokens": 5000 },
+  "seats": {
+    "criteria": { "provider": "openrouter", "model": "meta-llama/llama-4-maverick", "maxTokens": 36000, "lab": "llama4-maverick" },
+    "skeleton": { "provider": "external", "model": "claude-code-session" },
+    "builder": { "provider": "external", "model": "claude-code-session" },
+    "reviser": { "provider": "external", "model": "claude-code-session" },
+    "handoff": { "provider": "external", "model": "claude-code-session" },
+    "proposers": [
+      { "provider": "openrouter", "model": "tencent/hy4-preview", "maxTokens": 36000, "lab": "hy4-preview" },
+      { "provider": "openrouter", "model": "google/gemini-3.8-flash", "maxTokens": 36000, "lab": "gemini3.8-flash" },
+      { "provider": "openrouter", "model": "deepseek/deepseek-v4.1-flash", "maxTokens": 36000, "lab": "deepseek-v4.1-flash" },
+      { "provider": "openrouter", "model": "z-ai/glm-5.2", "maxTokens": 36000, "lab": "glm5.2" },
+      { "provider": "openrouter", "model": "meta/muse-spark-1.3", "maxTokens": 36000, "lab": "muse-spark1.3" },
+      { "provider": "openrouter", "model": "qwen/qwen3.8-max-0902", "maxTokens": 36000, "lab": "qwen3.8-max" },
+      { "provider": "openrouter", "model": "mistralai/mistral-medium-3-5", "maxTokens": 36000, "lab": "mistral-medium3.5" }
+    ],
+    "alternatives": [
+      { "provider": "external", "model": "subscription:opus-5.5", "lab": "opus5.5-sub" },
+      { "provider": "openrouter", "model": "openai/gpt-6-luna", "maxTokens": 36000, "lab": "gpt6-luna" },
+      { "provider": "openrouter", "model": "deepseek/deepseek-v4-pro", "maxTokens": 64000, "lab": "deepseek-v4-pro" }
+    ],
+    "critics": [
+      { "provider": "external", "model": "subscription:opus-5.5", "lab": "opus5.5-sub" },
+      { "provider": "openrouter", "model": "openai/gpt-6-luna", "maxTokens": 36000, "lab": "gpt6-luna" },
+      { "provider": "openrouter", "model": "deepseek/deepseek-v4-pro", "maxTokens": 64000, "lab": "deepseek-v4-pro" }
+    ],
+    "deep_dive": { "provider": "openrouter", "model": "deepseek/deepseek-v4.1-flash", "maxTokens": 36000, "lab": "deepseek-v4.1-flash" }
+  }
+}

package/chains/plan-highest-7.json

@@ -0,0 +1,45 @@
+{
+  "name": "plan-highest-7",
+  "description": "Tiered council, Muad's design (2026-09-26): the highest models for the highest tasks, the cheapest models for the mass work. MASS SEATS (seats.proposers, the cheap-7-v2 roster: GPT-5.6 Luna, Gemini 3.8 Flash, DeepSeek V4.1 Flash, GLM 5.3 Flash, Qwen 3.7 Plus, MiniMax M3, Mistral Medium 3.5) write the blind proposals and all the proposal debate posts and replies, and post on the anchors' architectures (alternatives.debaters \"all\"). ANCHOR SEATS: GPT-6 Astra and Claude Opus 5.5 (Muad's pick 2026-09-26, in place of Fable 5.1; still his deliberate exception: an Anthropic model reviews a draft a Claude Code session writes), DeepSeek V4 Pro and GLM-5.3 write the whole alternative architectures (seats.alternatives), answer the posts on them, and form the review panel that must sign off unanimously (seats.critics). The mass seats argue; only the anchors vote, so a cheap majority can never out-vote them. DEEP-DIVE SEAT (seats.deep_dive, DeepSeek V4.1 Flash, a mass-tier model: it must not share a lab with a voting anchor, chain-lint deep-dive-votes): one job, checking the first draft against every part of the task, 40,000 characters per call at a 36k output cap, up to 24 calls under its OWN $4 cap inside the run's; it does not vote, and the reviser answers its findings before the anchors' first review. The MAJORITY GUARD is ON (majority_guard, Muad's go 2026-09-26; off in every other shipped chain but mock-tiered): authors see each argument once with no lab letter or count, the reply prompts carry an evidence bar, and a withdrawal that does not quote the argument it concedes to is kept for the builder. Criteria move to Llama 4 Maverick, which holds no other seat. Same Claude Code writer seats and stages as plan-premium-7, including decision records and the dispute stage. No Kimi, no xAI. Untested with real models: price it with `npm run dry` and get an explicit go before any paid run.",
+  "maxRounds": 7,
+  "signoff": "unanimous",
+  "dispute": { "enabled": true, "stall_rounds": 2 },
+  "proposals": { "parts": 3, "maxTokens": 4000 },
+  "debate": true,
+  "decisions": { "enabled": true },
+  "alternatives": { "enabled": true, "debaters": "all" },
+  "deep_dive": { "enabled": true, "job": "sources", "usd": 4, "maxCalls": 24, "chunkChars": 40000 },
+  "majority_guard": { "enabled": true },
+  "canary": { "enabled": true, "sampleRate": 0.1 },
+  "handoff": true,
+  "estimate": { "promptTokens": 10000, "draftTokens": 10000, "critiqueTokens": 5000 },
+  "seats": {
+    "criteria": { "provider": "openrouter", "model": "meta-llama/llama-4-maverick", "maxTokens": 36000, "lab": "llama4-maverick" },
+    "skeleton": { "provider": "external", "model": "claude-code-session" },
+    "builder": { "provider": "external", "model": "claude-code-session" },
+    "reviser": { "provider": "external", "model": "claude-code-session" },
+    "handoff": { "provider": "external", "model": "claude-code-session" },
+    "proposers": [
+      { "provider": "openrouter", "model": "openai/gpt-5.6-luna", "maxTokens": 36000, "lab": "gpt5.6-luna" },
+      { "provider": "openrouter", "model": "google/gemini-3.8-flash", "maxTokens": 36000, "lab": "gemini3.8-flash" },
+      { "provider": "openrouter", "model": "deepseek/deepseek-v4.1-flash", "maxTokens": 36000, "lab": "deepseek-v4.1-flash" },
+      { "provider": "openrouter", "model": "z-ai/glm-5.3-flash", "maxTokens": 36000, "lab": "glm5.3-flash" },
+      { "provider": "openrouter", "model": "qwen/qwen3.7-plus", "maxTokens": 36000, "lab": "qwen3.7-plus" },
+      { "provider": "openrouter", "model": "minimax/minimax-m3", "maxTokens": 36000, "lab": "minimax-m3" },
+      { "provider": "openrouter", "model": "mistralai/mistral-medium-3-5", "maxTokens": 36000, "lab": "mistral-medium3.5" }
+    ],
+    "alternatives": [
+      { "provider": "openrouter", "model": "openai/gpt-6-astra", "maxTokens": 36000, "lab": "gpt6-astra" },
+      { "provider": "openrouter", "model": "anthropic/claude-opus-5.5", "maxTokens": 36000, "lab": "opus5.5", "estimate": { "critiqueTokens": 10000 } },
+      { "provider": "openrouter", "model": "deepseek/deepseek-v4-pro", "maxTokens": 64000, "lab": "deepseek-v4-pro" },
+      { "provider": "openrouter", "model": "z-ai/glm-5.3", "maxTokens": 64000, "lab": "glm5.3" }
+    ],
+    "critics": [
+      { "provider": "openrouter", "model": "openai/gpt-6-astra", "maxTokens": 36000, "lab": "gpt6-astra" },
+      { "provider": "openrouter", "model": "anthropic/claude-opus-5.5", "maxTokens": 36000, "lab": "opus5.5", "estimate": { "critiqueTokens": 10000 } },
+      { "provider": "openrouter", "model": "deepseek/deepseek-v4-pro", "maxTokens": 64000, "lab": "deepseek-v4-pro" },
+      { "provider": "openrouter", "model": "z-ai/glm-5.3", "maxTokens": 64000, "lab": "glm5.3" }
+    ],
+    "deep_dive": { "provider": "openrouter", "model": "deepseek/deepseek-v4.1-flash", "maxTokens": 36000, "lab": "deepseek-v4.1-flash" }
+  }
+}

package/chains/verify.json

@@ -1,6 +1,6 @@
 {
   "name": "verify",
-  "description": "The recommended default. Two labs, a hard 2-round cap: the smallest cross-lab review, and the cheapest.",
+  "description": "The CLI default when --chain is not given. Two labs, a hard 2-round cap: the smallest cross-lab review. For a first real council, start with cheap-7-v2 (one OpenRouter key); see the README's chain table.",
   "maxRounds": 2,
   "stopOnPass": true,
   "estimate": {

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "the-high-council",
-  "version": "0.7.7",
+  "version": "0.7.9",
   "mcpName": "io.github.muad-yasin/the-high-council",
   "description": "The High Council: a multi-model planning harness. One builder, independent critics from other labs, checkable acceptance criteria, a debate on the record, and a hard stop.",
   "type": "module",
@@ -49,11 +49,11 @@
   },
   "dependencies": {
     "@modelcontextprotocol/sdk": "^1.30.0",
-    "express": "^4.22.2",
     "zod": "^4.6.4"
   },
   "files": [
     "src/",
+    "!src/http-server.js",
     "chains/",
     "schemas/",
     "types/",
@@ -84,6 +84,7 @@
   ],
   "devDependencies": {
     "@yao-pkg/pkg": "6.22.0",
-    "ajv": "^8.20.0"
+    "ajv": "^8.20.0",
+    "express": "^4.22.2"
   }
 }